This invention relates generally to medical ultrasonic imaging, and more specifically to improved methods for transmitting pulses with desired power levels.
Brock-Fisher et al. (U.S. Pat. No. 5,577,505) describe several methods for isolating non-linear responses from contrast agent and tissue. In one method, multiple pulses are fired along a line, where one pulse uses the entire transducer aperture. The aperture is then divided into two or more subapertures, and other pulses fired along the same line using the subapertures. The echo signals produced by these other pulses are then combined and subtracted from the echo signal produced by the first pulse to determine the non-linear response.
Thomas et al. (U.S. patent application Ser. No. 09/514,803) describe an alternative technique, called xe2x80x9cContrast Pulse Sequences,xe2x80x9d (CPS), where multiple pulses are transmitted into the body. These pulses have different transmit amplitudes and phases. For example, three pulses can be transmitted with amplitudes +1, xe2x88x922, and +1. The three pulses are then summed on receive so that linear responses are cancelled, and non-linear responses are preserved.
In both approaches, different pulses with different amplitudes are used so that odd-order non-linearities are preserved. By contrast, pulse inversion techniques, such as described by Hwang et al. (U.S. Pat. No. 5,951,478) and Chapman et al. (U.S. Pat. No. 5,632,277), remove odd-order non-linearities because the amplitude of the transmit pulse does not change when the phase is inverted.
The technique described by Brock-Fisher et al. suffers from degraded sensitivity to second harmonic signals. Consider the case where three pulses are fired along an acoustic line. The first pulse is fired using the entire aperture; the second pulse is fired using the even elements of the aperture; and the third pulse is fired using the odd elements of the aperture. Assume that a target is located at the transmit focus of the aperture along the acoustic line. In this case, the amplitude from each of the two half-aperture firings will be about one half the amplitude of the full-aperture firing.
If the pressure at the target from the half-aperture firing is one half the amplitude of the full-aperture firing, and if we assume that the second harmonic response of the signal is proportional to the square of the pressure, then the second harmonic response from the half-aperture is one quarter that of the full-aperture firing. Since the half-apertures are fired twice and summed, the total second harmonic response from the half-apertures will be one half that of the full-aperture firing. In this technique, the sum of the responses from the half-apertures is then subtracted from the response from the full aperture, resulting in a second harmonic signal with an amplitude one half that of the full aperture firing by itself. Such a loss in signal may be unacceptable, especially since the non-linear signals may be weak to begin with.
This problem is overcome by the Contrast Pulse Sequences described by Thomas et al. In one example, three pulses are fired with amplitudes +xc2xd, xe2x88x921, +xc2xd, and the resulting echo signals are summed on receive. Note the big difference between these sequences, and the sequences described by Brock-Fischer, is that the transmit phase changes with different firings. The second harmonic echo signals for these transmit pulses will be xc2xc, 1, and xc2xc, respectively. By summing these echo signals, the second-harmonic response will have an amplitude of 1.5. Note that the amplitude of the second harmonic signal is greater than the second harmonic response from any individual firing. Thus, the second harmonic signal is preserved while the fundamental signal is cancelled.
The CPS technique relies on the precise control of amplitude of the transmitted pulses. Non-linear elements of the transmitter circuitry and the transducer can make this quite difficult to achieve. For example, diodes in the transmitter circuitry may cause problems since they often result in a fixed voltage drop. Attempts to correct for these non-linearities can be quite difficult, since these non-linearities can vary with different transmit voltages.
One preferred embodiment described below addresses this problem by maintaining the same power level for each transducer element throughout a set of pulses that includes at least two pulses of different amplitude and at least two pulses of different phase. In this embodiment, the larger-amplitude pulse uses a larger aperture including a selected number of transducer elements. The smaller-amplitude pulses of different phase use smaller apertures, selected such that each of transducer elements that is active in the larger-amplitude pulse is active in one or the other (but not both) of the lower-amplitude pulses. In this way each transducer element is active for the same number of pulses at the same power level for both the larger-amplitude pulse or pulses and the smaller-amplitude pulse or pulses. This ensures that the echo signals from the pulse sequence, when summed, have the desired amplitudes to cancel the selected fundamental and/or harmonic frequencies.
Because the voltage applied to each transducer element is held constant when that transducer element is active, problems related to an ultrasound system""s transmit non-linearities are to a large extent solved. All odd-order non-linearities are removed. Though even-order non-linearities of the transmitter or transducer will still exist, most ultrasound systems are designed to have very small even-order non-linearities on transmit. For example, non-linear voltage-to-pressure responses will not adversely affect pulse amplitude. Since the phase of the smaller-amplitude pulses is different from the phase of the larger-amplitude pulse, the response from second order non-linearities will be greatly improved over the technique described by Brock-Fisher.
This section has been provided by way of general introduction, and is not intended to limit the scope of the following claims.